Академический Документы
Профессиональный Документы
Культура Документы
Volume 1 | Issue 2
January 2012
Lorna Lueck
University of Massachusetts, Amherst, lolueck@web.de
Erik B. Erhardt
University of New Mexico, erike@stat.unm.edu
Lyle E. Craker
University of Massachusetts, Amherst, craker@umass.edu
Recommended Citation
Gardner, Zoe E.; Lorna Lueck; Erik B. Erhardt; and Lyle E. Craker. 2012. "A morphometric analysis of Actaea racemosa L.
(Ranunculaceae)." Journal of Medicinally Active Plants 1, (2):47-59.
DOI: 10.7275/R5M906KB
http://scholarworks.umass.edu/jmap/vol1/iss2/3
This Article is brought to you for free and open access by ScholarWorks@UMass Amherst. It has been accepted for inclusion in Journal of Medicinally
Active Plants by an authorized editor of ScholarWorks@UMass Amherst. For more information, please contact scholarworks@library.umass.edu.
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
June 2012
Lorna Lueck
University of Massachusetts, Amherst, lolueck@web.de
Erik B. Erhardt
University of New Mexico, erike@stat.unm.edu
Lyle E. Craker
University of Massachusetts, Amherst, craker@umass.edu
Recommended Citation
Gardner, Zoe E., Lorna Lueck, Erik B. Erhardt, Lyle E. Craker. 2012. "A morphometric analysis of Actaea racemosa L.
(Ranunculaceae)," Journal of Medicinally Active Plants 1(2):47-59.
DOI: https://doi.org/10.7275/R5M906KB
Available at: http://scholarworks.umass.edu/jmap/vol1/iss2/3
This Article is brought to you for free and open access by ScholarWorks@UMass Amherst. It has been accepted for inclusion in Journal of Medicinally
Active Plants by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact
scholarworks@library.umass.edu.
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
1
Department of Plant, Soil & Insect Sciences, University of Massachusetts, Amherst, MA 01003 U.S.A.
2
Department of Mathematics and Statistics, MSC01 1115, 1 University of New Mexico, Albuquerque, NM
87131 U.S.A.
Keywords: Cimicifuga racemosa, medicinal plant, conservation, morphology, morphometrics, plant geography,
Tukey-Kramer multiple comparisons, UPGMA cluster analysis
47
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
488
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
Morphological analysis. In each population, primary vein. A leaflet was considered adhering if
20 plants were examined and morphological data on more than three mm of the base of the leaflet was
the height of the mature plant, height of the main fused with the petiolule.
compound leaf, number of compound leaves, length
of the terminal leaflet, number of inflorescences,
height of inflorescences, and the stage of repro-
duction was recorded at the time of collection (Figure
2). From each plant, the three terminal leaflets of the
largest leaf and five flowers were collected. If the
desired leaflets were missing or severely deformed,
botanically equivalent leaflets, usually from a side
branch of the same compound leaf, were collected.
Such substitution occurred in approximately 2.5
percent of plants sampled. During collection, the
flowers were picked from each flowering plant,
immediately pressed flat and allowed to dry in paper
envelopes stored in silica gel. Sets of leaflets were
picked and kept in a plastic bag until being pressed in
newsprint, three to nine hours after collection. The
botanical identity of collected plants was verified at
the time of collection through the observation of
reproductive and vegetative parts and was later
confirmed by AFLP fingerprinting (Lueck, 2003).
Based on previous work in this genus
(Compton and Hedderson, 1997; Compton, Culham,
and Jury, 1998; Lee and Park, 1994; Ramsey, 1987) Figure 2. Morphological measurements of plant parts.
and initial examination of characteristics that A=whole plant, B=leaflets, C=flowers, D= staminodia.
appeared to vary between populations, 13 lengths and
3 angles were measured on each leaflet. Leaflets
were measured with the image analysis program
Scion Image Beta 4.0.2 (Scion Corp., Frederick,
MD). Pressed leaflets were scanned (HP ScanJet
6200) to create digitized images of the leaflets and
the digitized images were calibrated using the
scanned image of a millimeter grid scanned with each
leaflet.
Selected lengths (in cm) and angles (in degrees)
in the images were measured using measurement
tools in the Scion program. Characteristics of the
secondary leaflets were categorized into one of the
following five categories: 1) petiolule present, base
meeting; 2) petiolule present, base oblique; 3) Figure 3. Lateral leaf base categories.
petiolule absent, base meeting; 4) petiolule absent,
base oblique; 5) petiolule absent, base adhering (Figure Of the 26 populations sampled, 16 contained
3). A leaflet base was considered oblique if the base flowering plants. Single flowers of five plants per
on one side of the primary vein was more than three population were examined. Dried flowers were rehy-
millimeters from the base on the opposite side of the drated for a minimum of 10 min in 70% ethanol. At
49
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
the time of examination, each flower was placed on Statistical analysis. Statistical analyses were
a Petri dish and several milliliters of the ethanol done using statistical software packages SAS Release
solution were added to keep the flower hydrated and 8.00 (SAS Institute Inc., Cary, NC) or Minitab
easy to manipulate. Under a 10X binocular dis- Release 14.20 (Minitab, Inc., State College, PA).
secting scope, the lengths of the flower bract, Unless otherwise indicated, statistics displayed are
pedicel, stamens, and pistil were measured using for all individuals (rather than population averages of
digital calipers (Mitutoyo Plastical digital calipers) individuals). Descriptive statistics were generated,
and the number of stamens and staminodia were listing the minimum, maximum, mean, standard
counted. Staminodia were removed with a dissect- deviation, and coefficient of variation for each
ing needle and stored in 70% ethanol until further characteristic. For each quantitative characteristic an
examined. Staminodia were placed on a glass slide analysis of variance (ANOVA) test for equal
with several drops of ethanol solution. A graticule population means was run, followed with the Tukey-
in the eyepiece of a 40X binocular microscope was Kramer multiple comparison test (HSD) for unequal
used to measure selected dimensions on each stami- sample sizes to indicate pairwise differences of
nodium and the apex and base characteristics of each population means using SAS (Kramer, 1956; Tukey,
staminodium were scored (Figure 4). 1953).
Dendrograms were created based on different
selections of the available data using the "cluster ob-
servations" command in Minitab 14, using the
UPGMA algorithm on standardized variables with
average linkage and squared Euclidean distances
(Lance and Williams, 1967). For consensus, dendro-
grams were constructed using combinations of linkage,
distance, and standardized and unstandardized vari-
ables and results were consistent.
In total three datasets were used to create
dendrograms: (1) all populations using averages of
non-flower characteristics (characteristics labeled 1-
23, (2) populations with floral data using all available
data (characteristics labeled 1-37) with plant averages
of staminodium characteristics (characteristics labeled
31-37) and (3) populations with floral data using floral
data only (characteristics 24-37) with plant averages of
staminodium characteristics (characteristics 31-37).
RESULTS
An overview of variation for all characteristics
measured in the study was established (Table 2) with
ANOVA F-statistics indicating the presence of sig-
nificant differences between at least two populations.
For 23 of 37 characteristics, statistical differences
between populations were indicated. For the character-
istics 7, 9-11, 16, 22, 24, 26, 28-31 and 35-37, no differ-
ences between populations were observed. Tukey--
Figure 4. Staminodium apex and base types. Kramer testing of individual characteristics provided
groupings that indicate significant pairwise pop-
ulation differences (Table 3).
50
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
3
Whole plant characteristics
17. Height (cm) 450 20 45 248 154.07 32.18 20.89 11.39
18. Tallest leaf height (cm) 511 20 27 90 53.60 10.41 19.42 18.76
19. Compound leaves (number) 511 20 1 6 2.21 0.91 41.18 N/A
20. Length three terminal leaflets (cm) 511 20 11 31 19.27 3.48 18.06 8.24
4
21. Length largest inflorescence (cm) 450 20 17 89 29.67 9.48 31.95 11.65
22. Flower stalks (number) 511 20 0 3 0.38 0.51 134.21 N/A
23. Inflorescences (number) 511 20 0 17 2.70 1.98 73.33 17.31
5
Flower characteristics
24. Bract length (mm) 36 15 1.0 5.6 2.66 0.56 21.05 1.31
25. Pedicel (mm) 83 17 2.7 8.2 5.07 1.09 21.50 4.57
26. Stamen length (mm) 83 17 3.6 8.5 6.25 0.99 15.84 2.45
27. Stamens (number) 83 17 43.0 134.0 95.37 15.41 16.16 3.09
28. Pistil length (mm) 83 17 1.7 5.3 3.61 0.49 13.57 2.82
29. Staminodium length (mm) 83 17 2.4 4.7 3.26 0.45 13.80 0.58
30. Staminodia (number) 83 17 8.0 4.4 0 1.73 39.32 1.81
31. Staminodium width top (mm) 350 16 1.0 12.5 6.33 2.07 32.65 3.08
32. Staminodium width midpoint (mm) 350 16 3.0 10.0 5.85 1.15 19.73 11.52
33. Staminodium length top (mm) 350 16 1.0 16.0 5.34 1.80 33.62 7.30
34. Staminodium length midsection (mm) 350 16 5.0 23.5 13.28 2.50 18.83 8.71
35. Staminodium length base (mm) 350 16 5.0 34.0 12.98 3.38 26.02 2.94
36. Apex type (score) 350 16 N/A N/A N/A N/A N/A N/A
37. Base type (score) 350 16 N/A N/A N/A N/A N/A N/A
1
Coefficient of variation is a percentage value of the standard deviation divided by the mean.
2
The between versus within population variation. Values larger than about 3 indicate significant differences between at least two
population means.
3
Whole plant characteristics were recorded for all plants. A total of 61 non-flowering plants were sampled and no total height or
inflorescence length was recorded for these plants.
4
The shortest length in an inflorescence beyond BBCH stage 60 (first flowers open) (Bleiholder et al., 1997).
5
A total of 83 flowers (collectively having 350 staminodia) were examined. Bracts were separated from many of the dried flowers,
but remained attached to 36 flowers (characteristic 24).
N/A = Not applicable.
51
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
Whole plant morphology. The largest popu- leaflet base. While all terminal leaflets had petio-
lation mean was roughly twice that of the smallest lules, petiolules were present on only 20 percent of
population mean for all leaflet characters and less lateral leaflets.
than twice for flower characteristics. The tallest in- Flower morphology. Flower morphology was
dividual plants were observed in populations labeled variable both within and between populations for all
NY-2, IN-1, and WV-3). These populations also had characteristics examined. Certain characteristics
large numbers of inflorescences, the greatest leaf demonstrated a relatively large amount of variation
height, more than the average number of leaves, and between population means and statistically significant
relatively long inflorescences. differences could be observed between populations.
Leaflet morphology. Coefficients of variation These include pedicel length, staminodium midsec-
for leaflet characteristics ranged from 17.75 to 51.43. tion width, staminodium tip length, and staminodium
The Tukey-Kramer analysis revealed large, over- midsection length. Other characteristics demonstrated
lapping groups of populations with similar ranges. more uniformity among populations and the popu-
Only populations near the minimum and maximum lations did not differ in flower morphology.
for certain characteristics were significantly different The number of stamens per flower ranged be-
from each other. For instance, populations MD-1, tween 43 and 134. The smallest variation in the num-
SC-1, PA-2 and MA-1 had several particularly large ber of stamens was in population KY-2, with stamen
leaflet characteristics, while populations NC-6, NC-5, numbers ranging from 71 to 74, population PA-2 had
VA-1 and MO-1 were smaller. the largest range, 43 to 114, and population SC-1 had
Specifically, populations NC-6 and MO-1 had the highest mean number of stamens (90 to 134 per
smaller than average leaflet characteristics, including flower).
terminal leaflet length, terminal leaflet width, middle Bracts, present on 40 of the 88 flowers examin-
lobe length, and middle lobe width at base. Co- ed, ranged from 1 to 5.6 mm in length. Although all
efficients of variation in these characteristics in these populations were statistically equivalent, population
populations were generally smaller than the variation WV-2 had both the longest bract and largest vari-
observed in other populations. Population MD-1 had ability. Staminodia demonstrated both within and be-
the largest average measurements for many leaflet tween population variability, similar to that of other
characteristics, including the terminal leaflet length, floral traits (Figure 5).
terminal leaflet width, middle lobe length of the ter- Distinctive populations within traits included
minal leaflet, lateral lobe length of the terminal population NC-2 with large staminodium width at
leaflet, lateral lobe angle relative to the vertical axis midpoint, population NC-3 with small and relatively
and lateral leaflet length. uniform staminodium midsection length, and popu-
Certain characteristics demonstrated a relatively lations WV-2 and WV-3 with relatively large varia-
large amount of variation, as indicated by the Tukey- tion in staminodium base length. The shapes of the
Kramer analysis for a number of present groupings, staminodium apices and bases demonstrated a
including terminal leaflet length, terminal leaflet surprising amount of plasticity as compared with the
width, middle lobe width at base, length of terminal summary by Ramsey (1987) that concluded stamina-
leaflet base to lateral lobe apex, lateral leaflet length, dia shapes are stable within the species.
lateral leaflet width at midpoint, and angle of terminal In total, six different types of staminodia apices
leaflet apex. Other characteristics demonstrated a were recognized. Most populations shared bifid apices
greater amount of uniformity of population means that branched into two narrow lobes of similar length,
between populations, including length of middle lobe but variable form. In addition, unusual types occurred
on terminal leaflet, middle lobe width at midpoint, where the two lobes were nearly or entirely merged,
length of lateral lobe on terminal leaflet, lateral lobe or where the two lobes were enlarged into oval struc-
angle relative to vertical axis, lateral lobe width at tures. Populations KY-1, MO-1, and WV-2 were
base, lateral lobe width at midpoint, length of most variable in the staminodium apices because all
petiolule of terminal leaflet and angle of terminal six types were present. The populations NC-2, NY-2,
52
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
and SC-1 appeared most homogenous in this trait however, some patterns of morphological variation
because only three apex types were observed in these appear possible.
populations. Merged lobes were only observed in 9 of While the observed variation in this study did
the 17 populations, while enlarged lobes were present not allow delineation of groups based on leaflet
in even fewer populations (KY-1, MO-1, NY-1 and morphology, some variation was noted in the dif-
WV-2). ferent leaflet characteristics, enabling the discernment
of certain populations for selected traits. Levels of
phenotypic variation detected depend on the
characteristics measured and more variation may be
expected in leaves than in flowers (Lawrence, 1950;
Stace, 1989). In this study, the coefficients of varia-
tion for leaflet characteristics observed are generally
higher than those observed for flower characteristics.
Means of characteristics in this study are similar
to those reported by others examining A. racemosa.
Ramsey (1987) reported a mean terminal leaflet
length of 10.5 cm for A. racemosa, and a mean
terminal leaflet width of 8.1 cm, as compared to the
10.4 cm terminal leaflet mean length and 7.4 cm
mean width in this study. Compton (1982) reports
the number of staminodia as 1 to 8, as compared to
Figure 5.Occurrence of staminodium apex type.
our 0 to 8, and stamens as 55 to 110 as compared
with our 43 to 134.
Relationship among populations. Multivariate
Commenting on staminodium morphology,
summaries of population similarity are illustrated by
Ramsey (1987) noted that staminodia shapes were
a dendrogram (Figure 6). In the dendrogram, the
stable within species. The variation observed in
emerging patterns and groups do not correspond with
staminodium characteristics in this analysis was
geographic location or altitude. The groupings de-
much greater than anticipated and greater than
pend strongly on the data subsets used and may be
reported by Lee and Park (1994) in A. foetida and
the reason dendrograms created on different data sets
Ramsey (1987) in North American species of Actaea.
do not reveal similar patterns. Using vegetative data,
This greater variation is surprising given that
populations with a small leaflet width, MO-1, NC-6,
morphological diversity appears to be higher in A.
and NY-1, form a distinct cluster.
foetida than A. racemosa (Compton and Hedderson
1997).
DISCUSSION
Ramsey (1965), studying 2000 herbarium
Several species of Actaea grow in the eastern
specimens of A. racemosa, found 16 unique speci-
United States. These species are recognized as being
mens labeled as dissecta, a teratological form of the
closely related, suggesting a relatively recent evolu-
species that has highly dissected leaflets. In sampling
tionary division (Compton 1982; Ramsey 1986,
a set of populations in this study that cover a signifi-
1988) and reducing the likelihood of within species
cant portion of the geographical range of this species,
differentiation. Even different species are difficult to
no such unique individuals or groups that could be
distinguish when the plants are not in flower, as the
classified into forms were observed. This lack of
leaflet morphology of the different species is very
unique individuals is not surprising, as the analysis b
similar (Ramsey 1965). With this level of similarity,
Ramsey (1965) included many herbarium specimens
elucidating groups with typical morphological traits
that likely represented a significantly higher propor-
below the species level can be challenging. Given the
tion of the unusual forms than would be found in wild
wide geographical range of the sampled populations,
populations.
53
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
While A. racemosa is typically a plant of decid- with subsequently lower morphological differentia-
uous woodlands and most populations observed in tion among populations could be expected (Hamrick
this study were growing alongside typical woodland &Godt 1989).
understory plants as Sanguinaria canadensis L.,
Asarum canadense L., Polystichum acrostichoides CONCLUSIONS
(Michx.) Fe., Adiantum pedatum L., Impatiens This study assessed the morphological variation
pallida Nutt., and Arisaema triphyllum (L.) Schott, A. of A. racemosa to identify patterns of variation at the
racemosa was also observed and collected from population and species levels. While variation was
atypical sites, such as a hillside clearing with no can- observed for all characteristics, cluster analyses
opy cover and alongside Phragmites australis (Cav.) indicated morphological variation within populations
Trin., Achillea millefolium L., and Verbascum was similar to that between populations and that this
Thapsus Bertol (population NY2). This observation variation was not influenced by geographical
illustrates the adaptability of the species to different distribution.
growing conditions and the variability of habitats in While no unique phenotypes were observed,
which A. racemosa grows. discernment of some populations based on leaf and
The sampling protocol was designed to exclude flower characteristics was possible, suggesting a
variation due to different development stages, while starting point the development of possible
variability due to different microclimates within morphologically defined and homogenous cultivars.
populations was not excluded. Given the large size of
some populations, the protocol procedures enabled ACKNOWLEDGEMENTS
insight on the magnitude of variation within This material is based on work supported by
populations. While the variation makes differentia- generous funding from the German Leopoldina
tion of populations according to morphological traits Akademie der Naturforscher with funds from the
quite difficult, the naturally occurring variability of German Federal Ministry of Education and Research
the species is reflected. The observed adaptability (grant number BMBF-LPD 9901/8-58) and the
makes A. racemosa more amenable to cultivation Cooperative State Research Experiment Station and
than other woodland medicinal plant species, such as the Department of Plant, Soil, & Insect Sciences
Panax quinquefolius L. and Hydrastis canadensis L. (paper number 3430 under Project No. MAS000729).
In addition to ecological variability, the plant The authors further thank the following for their
breeding system can influence genetic differentiation support of this research: U.S. Forest Service, U.S.
and cause subsequent morphological differentiation Natural Heritage Network, Yellow Creek Botanical
among populations. A. racemosa is a slowly repro- Institute, United Plant Savers, The Triangle Land
ducing (Baskin and Baskin 1985) and slowly Conservancy, Dr. Joe-Ann McCoy, Ms. Megan
migrating (Matlack, 1994) species, suggesting that Peabody, Dr. Karen Searcy, Dr. James Walker, Russ
differentiation do to distance between populations Richardson, Gary Kauffman, Eric Burkhard, Chip
should be possible. The species, however, is a long- Carroll, Dr. Scott Mori, Dr. Allison Miller, Dr.
lived perennial, pollinated by insects and by pollen- Gwynn Ramsey, Bill McAvoy, Robin Suggs, and the
ovule ratios averaging over 30000:1 (unpublished other generous volunteers who assisted in locating the
data), which indicate, according to Cruden (1977), A. plant populations. Detailed comments from an anon-
racemosa is most likely xenogamous. Based on the ymous reviewer helped to improve this manuscript
species longevity, wide distribution, large population and are gratefully acknowledged.
sizes, and outcrossing characteristics, gene flow be-
tween populations and lower genetic differentiation
54
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
5. Middle lobe width at midpoint (cm) 6. Terminal leaflet length (cm) 12. Lateral leaflet length (cm) 13. Lateral lobe width at midpoint (cm)
Pop Grouping Mean N Pop Grouping Mean N Pop Grouping Mean N Pop Grouping Mean N
VA2 A 3.24 20 SC1 A 9.39 20 MD1 A 11.13 20 SC1 A 4.28 20
MD1 B A 3.19 20 MD1 B A 9.02 20 SC1 B A 11.08 20 DE2 B A 4.07 20
DE2 B AC 3.13 20 WV3 B A 8.86 20 MA1 B A 11.05 12 MD1 B A 4.00 20
PA2 BDAC 2.80 20 MA1 B AC 8.78 12 WV3 B A C 10.93 20 MA1 B AC 3.91 12
OH2 BDAC 2.62 20 PA2 BDAC 8.56 20 NC1 BDA C 10.90 20 WV3 BDAC 3.81 20
WV3 BDAC 2.61 20 TN1 EBDAC 8.28 20 PA2 EBDA C 10.44 20 KY2 EBDAC 3.58 20
VA1 EBDAC 2.50 20 DE2 EBDACF 8.24 20 KY1 EBDA C 10.38 20 NC1 EBDAC 3.48 20
DE1 EBDAC 2.50 20 VA2 EBDACF 8.18 20 NC2 EBDA C 10.06 20 PA2 EBDAC 3.43 20
MA1 EBDAC 2.47 12 NC1 EBDACF 8.15 20 TN1 EBDA C 9.92 20 WV2 EBDFC 3.38 20
SC1 EBDAC 2.41 20 NC2 EBDACF 8.10 20 DE2 EBDA CF 9.87 20 VA2 EBDFC 3.38 20
IN1 EBDACF 2.38 20 KY1 EBDACF 8.07 20 VA2 EBDA CF 9.80 20 DE1 EBDFC 3.36 20
NY2 EBDACF 2.38 20 NY2 EBD CF 7.75 20 WV2 EBDAGCF 9.60 20 TN1 EBDFC 3.28 20
KY2 EBDACF 2.36 20 WV2 EBD CF 7.70 20 NY2 EBDAGCF 9.59 20 VA1 EBDFC 3.28 20
OH1 EBDACF 2.26 20 WV1 EBD CF 7.66 20 IN1 EBDAGCF 9.52 20 KY1 EBDFCG 3.24 20
WV2 EBDACF 2.26 20 DE2 EBD CF 7.62 20 NC3 EBDAGCF 9.44 20 OH1 EBDFCG 3.23 20
NC1 EBD CF 2.25 20 OH2 EBDGCF 7.58 20 KY2 EBD GCF 9.40 20 NY2 E DFCG 3.15 20
KY1 E D CF 2.14 20 NC3 EBDGCF 7.57 20 OH1 EBD GCF 9.39 20 OH2 E DFCG 3.14 20
PA1 E D F 2.10 20 KY2 E DGCF 7.36 20 DE1 E D GCF 9.29 20 IN1 E DFCG 3.09 20
NC5 E D F 2.06 17 OH1 E DGCF 7.35 20 OH2 E D G F 9.20 20 NC2 E DF G 3.05 20
MO1 E D F 1.97 20 PA1 E DGCF 7.32 20 WV1 E G F 9.15 20 PA1 E DF G 2.98 20
WV1 E D F 1.94 20 NY1 E DG F 7.26 20 PA1 E G F 8.97 20 WV1 E DF G 2.98 20
NC3 E D F 1.90 20 IN1 E DG F 7.15 20 NY1 E G F 8.86 20 NC5 E F G 2.95 17
NC2 E D F 1.90 20 VA1 EH G F 6.88 20 NC5 E G F 8.83 17 NC3 EH F G 2.83 20
TN1 E D F 1.88 20 NC5 H G F 6.78 17 VA1 HG F 8.19 20 NY1 H F G 2.58 20
NY1 E F 1.62 20 MO1 6.09 H G 20 MO1 HG 7.91 20 MO1 H G 2.40 20
NC6 H 5.55 20 NC6 H 7.06 20 NC6 H 2.06 20
F Value = 11.40; Pr>F < 0.0001 F Value = 9.20; Pr>F < 0.0001 F Value = 9.46; Pr>F < 0.0001 F Value = 9.63; Pr>F < 0.0001
55
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
F Value = 4.26; Pr> F<0.0001 F Value = 10.04; Pr>F < 0.0001 F Value = 11.29; Pr>F < 0.0001 F Value =18.72; Pr>F < 0.0001
19. Number of compound leaves 20. Length three terminal leaflets (cm) 21. Length largest inflorescence (cm) 23. Number of inflorescences
Pop Grouping Mean N Pop Grouping Mean N Pop Grouping Mean N Pop Grouping Mean N
NY2 A 3.85 20 TN1 A 23.40 20 WV3 A 38.55 20 NY2 A 5.35 20
PA2 B A 3.10 20 WV3 B A 21.50 20 NY2 B A 37.60 20 WV3 B A 5.15 20
SC1 B C 2.70 20 MA1 B AC 21.25 12 NC3 B A C 36.75 20 SC1 B AC 4.50 20
OH1 B CD 2.55 20 MO1 B AC 21.15 20 NC1 BDA C 34.90 20 IN1 B AC 4.35 20
WV3 BECD 2.50 20 MD1 B AC 21.15 20 SC1 EBDA C 33.95 20 PA2 BDAC 4.30 20
NC1 FBECD 2.35 20 DE2 BDAC 20.40 20 KY1 EBDA C 33.15 20 WV1 BDEC 3.55 20
NC6 FBECD 2.20 20 NC2 BDAC 20.35 20 TN1 EBDA C 32.80 20 PA1 FDEC 3.35 20
IN1 FBECD 2.20 20 SC1 BDAC 19.95 20 PA2 EBDA C 32.50 20 VA1 FDEC 3.30 20
NC5 FBECD 2.18 17 KY1 BD C 19.60 20 WV1 EBDA CF 31.90 20 NC1 GFDEC 3.25 20
PA1 FBECD 2.15 20 OH1 BD C 19.35 20 KY2 EBDAGCF 30.45 20 OH2 GFDEC 3.00 20
TN1 FBECD 2.15 20 OH2 BD C 19.15 20 IN1 EBDAGCF 29.90 20 KY2 GFDECH 2.90 20
VA1 FBECD 2.15 20 NC3 BD C 19.00 20 NC5 EBDAGCF 29.35 17 OH1 GFDECH 2.80 20
DE1 F ECD 2.10 20 IN1 BDEC 18.95 20 VA1 EBDAGCF 29.05 20 MO1 GFDECH 2.80 20
MD1 F ECD 2.10 20 PA2 BDEC 18.80 20 MO1 EBDAGCF 28.55 20 MD1 GFDECH 2.80 20
MO1 F ECD 2.10 20 PA1 BDEC 18.80 20 MD1 EBDAGCF 28.55 20 KY1 GFDECH 2.75 20
NC3 F ECD 2.10 20 WV2 BDEC 18.45 20 NC6 EBDHGCF 27.65 20 NC3 GFDE H 2.55 20
NY1 F ECD 2.05 20 KY2 BDEC 18.45 20 OH1 EBDHGCF 27.60 20 NC6 GF E H 2.35 20
OH2 F ECD 2.05 20 WV1 BDEC 18.45 20 OH2 EBDHGCF 27.45 20 DE1 GF EIH 1.90 20
WV2 F ECD 2.05 20 DE1 BDEC 18.35 20 MA1 E DHGCF 26.67 3 NC5 GF IH 1.76 17
KY2 F ECD 2.05 20 VA1 BDEC 18.05 20 PA1 E DHGCF 26.35 20 TN1 GF IH 1.70 20
KY1 F ECD 2.00 20 NC5 BDEC 18.00 17 NY1 E DHG F 25.29 7 WV2 G IH 1.53 19
WV1 F ECD 1.95 20 NY2 DEC 17.90 20 WV2 E HG F 23.81 16 DE2 IH 1.20 20
DE2 F ECD 1.90 20 VA2 FDE 16.95 20 DE1 HG F 21.13 16 NY1 I 0.55 20
NC2 F E D 1.70 20 NC6 FDE 16.90 20 DE2 HG 20.06 16 MA1 I 0.50 12
VA2 F E 1.55 20 NC1 F E 15.45 20 VA2 H 17.29 7 VA2 I 0.45 20
MA1 F 1.42 12 NY1 F 14.20 20 NC2 I 3.29 7 NC2 I 0.35 20
F Value = 6.51; Pr>F < 0.0001 F Value = 8.24; Pr>F < 0.0001 F Value = 10.39; Pr>F < 0.0001 F Value = 17.37; Pr>F < 0.0001
56
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
F Value = 4.57; Pr>F < 0.0001 F Value =11.52; Pr>F < 0.0001 F Value = 7.30; Pr>F < 0.0001 F Value =8.71; Pr>F < 0.0001
a
For each characteristic, the Tukey-Kramer grouping indicates population means that are not significantly different at a 0.05 family-
wise significance level, the sample means in descending order, and the sample size; the ANOVA F-test statistic and p-value for the
test that all population means are equal is provided at the bottom of each table. Only characteristics for which the F value was .0001
or lower are shown here. All ANOVA tests were significant at the 0.05 significance level except for the characteristics bract length
and staminodium length A complete set for all characteristics is available upon request from the corresponding author.
57
Gardner et al.: A morphometric analysis of Actaea racemosa L. (Ranunculaceae)
59